The protein data bank (PDB) as a versatile resource for glycobiology and glycomics

Abstract

The compilation of data collections for carbohydrates has only recently gained momentum. The availability of such comprehensive databases, however, will be a prerequisite to successfully perform large-scale glycomics projects aiming to decipher new biological functions of glycans. With the carbohydrate structure suite (CSS), the carbohydrate-related data contained in the protein data bank (PDB) are now accessible through the Internet. It turned out that the PDB is a versatile resource for structural aspects in glycobiology. It provides reliable data about glycosylation sites, the conformational preferences of glycans and the specificity of protein carbohydrate recognition. A detailed comparison between the carbohydrate assignment reported in the PDB files and the nomenclature derived from atom coordinates guarantees that only consistent data will be evaluated. The automatic assignment of a unique structural description (LINUCS notation) enables easy cross-linking and referencing with other carbohydrate-related resources like NMR and MS-spectra. Exemplified for α- and β-N-acetylglucosamine it is shown that a particular distribution of amino acids is required to establish specific recognition for both anomers. The unrestricted use of primary data enables an online linkage of carbohydrate-related databases with other bioinformatics and biomedical resources and will thus provide maximal synergism.

• Carbohydrate
• glycobiology
• glycomics
• protein-carbohydrate interactions
• glycobioinformatics